Appendicies (these will be available here by 1 November 2002, or can be obtained from the Project Team - see contact details)
This paper sets out for consultation the exemplification of the proposals on the development of a new revenue grant allocation mechanism for Magistrates' Courts Committees in England and Wales.
The consultation is aimed at MCCs, Justices Chief Executives and other interested Magistrates' Courts Committee staff in England and Wales. This consultation is being conducted, as far as possible, in line with the Code of Practice on Written Consultation issued by the Cabinet Office.
Copies of the consultation paper are being sent to:
The Chairmen of the Magistrates' Courts Committees
Justices Chief Executives
Chief Executives of Lead Paying Authorities
Local Government Association
Members of Magistrates' Courts Consultative Council
HM Magistrates ' Courts Service Inspectorate
HM Treasury
Cabinet Office
Office of the Deputy Prime Minister
Wales Office
Welsh Assembly
Please send your response by 11th November 2003 to:
Gerry Callaghan Court Service
Resources and Planning Division
4th Floor
Selborne House
54-60 Victoria Street
London SW1E 6QW
Or by e-mail: gerry.callaghan@courtservice.gsi.gov.uk
The Department may wish to publish responses to this consultation document in due course. Please ensure your response is marked clearly if you wish your response or name to be kept confidential. Confidential responses will be included in any statistical summary of numbers of comments received and views expressed.
Further copies of this consultation paper can be obtained from Gerry Callaghan at the above address or by phoning 020-7210 0665.
Magistrates' Courts Committees (MCCs) are familiar with previous attempts to revise the grant allocation formula. The elements that constituted the 'frozen' formula cannot now be shown to correspond to MCCs' needs. It is very likely that funding is misallocated. This is potentially wasteful and undermines MCCs' ability to perform fully in accordance with the Government's agenda for criminal justice. The service has therefore accepted that the status quo cannot be maintained, despite the difficulty of moving to a new basis of allocation.
The principles underpinning the development of the proposed mechanism and its major components have already been the subject of two previous consultation exercises. The service has broadly accepted them, and they are not discussed again in this paper. The aim of this consultation paper is to exemplify a solution to this funding issue, using this year's allocation (2002/03) to give MCCs an understanding of how allocations may change under the new mechanism.
This year's allocation is used, first, because it is known and, secondly, because MCCs will be able to make a direct comparison between their actual allocations for 2002/03 and what they would have received if the new mechanism had been used then for grant allocation purposes.
The responses to the last Working Paper (WP 01/02) clearly indicated that most MCCs accepted both the need for change and the principles on which the proposed new mechanism is being established. However, they also showed that MCCs believe that the final acceptability of the mechanism will be determined by how it is actually implemented. It is for this reason that this paper seeks to demonstrate the behaviour of the mechanism under operational conditions.
This consultation paper seeks views on a number of detailed points. The main questions however relate to (1) whether MCCs understand how the mechanism produces their notional allocation and (2) their response.
MCCs may be inclined to answer question (2) in particular by whether their allocation increases or not. While this is understandable, MCCs will accept that local preferences alone cannot determine the question of the merits of the proposed formula for the funding of the whole service. Where consultees are not comfortable with the result, they should set out clear reasons for this.
Some of the responses may indicate that there is a difficulty about local data that has not been addressed, or some similar problem. Although the formal consultation period ends on 11 November 2002, MCCs are invited to contact the Project Team (see contact details at the beginning of this paper) as soon as they are aware of any such difficulty and the Project Team will aim to address it immediately.
During the consultation period, we also intend to issue a paper containing answers to frequently asked questions (FAQs). Consultees will also know that we plan to hold a conference on 31 October to support the consultation process and answer questions.
A brief overview of the responses to the Grant Allocation Mechanism Working Paper is presented below to identify the key themes arising from them.
The service is generally sympathetic to the aims of introducing the proposed grant allocation mechanism and broadly supports the theoretical principles on which it is based. A number of respondents mentioned that this was itself an improvement on previous suggested reforms, which had not been supported even at the theoretical level.
The main concerns about the outcome of the current work related to the:
This paper addresses most of these points, but some of them will also be dealt with in the FAQ paper (see paragraph 8 above).
The control total of grant that has been used for these exemplifications is the 2002/03 total grant allocation of £373,886,215.
For practical reasons which were discussed in Working Paper WP01/02, support and building costs are excluded at this stage in the development of the new grant allocation system. The responses to the working paper substantially supported this approach, at least in the short term. The unit process costs used in the allocation mechanism therefore cover only the operational expenditure of MCCs. Support and building costs are treated as fixed at the level of the last reported year's spending (2001/02) and have been identified in the cost model in line with the agreed guidelines.
Certain other costs are also excluded from the formula e.g. Libra costs and enforcement. These are referred to in the context of the overall allocation at paragraphs 44-48 below.
Some concerns have been expressed that top slicing support and building costs may have encouraged tactical behaviour by MCCs during the data collection exercise when allocating costs at the highest level between operations, support and buildings. However, the publication of agreed guidelines for making these allocations should have helped to minimise the likelihood of misallocations. In addition, all the costing models returned by MCCs have been scrutinised for differences in the interpretation of the guidelines and any apparent discrepancies followed up with the MCCs concerned and corrected if necessary.
The reported split of total costs across the operational, support and building costs for each MCC is presented in Appendix A.
The proportion of total costs identified as operational range from 57.5% for Lincolnshire to 82.8% for West Yorkshire, with an average of 71.5%. The proportion of total costs identified as support range from 6.8% for Northamptonshire to 32.0% for Lincolnshire, with an average of 19.3%. The proportion identified as building costs range from 4.6% for Gloucestershire to 14.8% for Nottinghamshire, with an average of 9.2%.
MCCs will note that the reallocation of grant between MCCs is driven by the unit costs of operational processes and not on the cost distribution across the major headings. Some of the MCCs that have reported the highest proportion of support service expenditure would lose grant through the implementation of the proposed new mechanism while some of those that have reported the smallest proportion of support expenditure would gain.
The distribution of operational expenditure across the various triggers is presented in Appendix B (i & ii).
MCCs will note that about 80% of their operational expenditure is accounted for by 5 main processes. For the future, this suggests that it may be possible to focus mainly on those processes e.g. for data collection purposes. Consideration will also be given to whether it is feasible to make next year's allocation on the basis of these main triggers alone.
At present there is no robust basis for forecasting future levels of activity for each process. In view of this, last year's activity levels (based on unweighted caseload) have been used to scale up the activity level data collected by each MCC over an eight week period and this has been used as the baseline activity level in the exemplifications. This approach was strongly supported in the responses to the Working Paper.
In order to ensure that the use of unweighted caseload data taken from existing computer systems was consistent, an audit exercise was carried out on the three main systems used by MCCs. This enabled adjustments to be made to counts to ensure consistency.
There may be advantages in using a three-year rolling average as the basis for determining the activity levels to be used in the formula in order to smooth out any annual fluctuations in activity levels. The practicality and desirability of pursuing this option will be investigated fully before the settlement is produced.
The exemplifications produced have used the top down costs for each process produced by each MCC. These are shown in Appendix C. Appendix C(i) shows process unit costs and C(ii) shows variances from the unit cost averages.
The process unit costs for each MCC exhibit the level of consistency that would be expected among a group of organisations such as MCCs. In other words there are some variations in unit costs for each process across the 42 but very few major outliers.
The Working Paper suggested the possible use of Standard Deviations (SDs) as an appropriate basis for this task and different SD-based options have been modelled. An explanation of the concept and use of SDs is presented in Appendix D. Appendix C shows the ± 1 SDs floors and ceilings for each of the 18 processes.
It was considered that there might be some value in exploring the potential for setting floors and ceilings for different "families" of MCCs. Rather than set thresholds on the basis of all MCCs, looking at more tightly defined clusters of MCCs that exhibit similar characteristics might be a way of introducing more discriminating floors and ceilings into the mechanism. However, neither the methodology for creating the families nor their use in an allocation mechanism is sufficiently developed at this stage for them to be considered for this purpose. They may prove a useful tool when looking at local cost differentials.
The basic mechanism set out in the working paper envisages the process costs of all those MCCs that lie within the floors and ceilings being allowed as measured for grant allocation purposes. Some of these MCCs may nevertheless still be inefficient while others will be underfunded. It is therefore for consideration whether direct steps should be taken at this stage to adjust the position of all MCCs, not just outliers.
However it has been suggested that the implementation of the new mechanism should not be too ambitious in the first instance and try to resolve every known limitation of the current system. The first step is to start with the major anomalies and work towards a more refined allocation across the board. For this reason these exemplifications focus only on the treatment of outliers and do not include any proposals for disallowing the actual unit costs for those MCCs that lie within the floors and ceilings. (Their allocations may nevertheless be affected by the redistribution of funding between those above and below the ceilings and floors.)
It is generally accepted that there are limits to the magnitude of changes in grant, either upwards or downwards, that MCCs can absorb from one year to the next. Something in the order of 3% is generally recognised as a sensible level at which to constrain year on year increases or decreases. However, there was also a view in favour of allowing larger increases in grant than decreases. An option allowing a 5% increase in grant as well as a 3% increase has therefore also been exemplified.
Although MCCs have already expressed their preference between the two options in response to Working Paper 01/02, they are invited to express a choice again if they wish in the light of the exemplifications.
As has already been noted above, the quality of data recorded for the purposes of the project is seen as critical to determining the acceptability of implementing the proposed approach. In view of this a data collection audit was undertaken jointly by the Internal Assurance Division of the Court Service and the Project Team and the resulting report has already been circulated to the service. In summary, although the report found a significant number of errors in the data collection processes based on spot field checks, the overall conclusion was that the data was fit for the purpose of supporting the grant allocation.
Validation work has also been undertaken on the top down cost models submitted by each MCC. In particular checks were made on the costs submitted to ensure that:
all costs were treated in accordance with the costing manual and subsequent technical releases;
all costs were correctly categorised between operational, support and building;
all costs were correctly allocated - i.e. apportioned or direct;
the spreadsheets added up i.e. that there were no omissions and no inclusion of extra costs;
the total costs agreed to the cost reconciliation; and
the total of all the triggers agreed to the total operational cost.
In addition the Project Team reviewed the costing work as it was in progress and identified errors to MCCs for correction or, in the later stages of the work, corrected the errors themselves and then notified the MCCs of any changes made.
However, data validation activities are still taking place for the bottom up costing exercise, including:
ensuring the bottom up total agrees with the top down;
ensuring that all costs are accounted for; and
seeking explanations for apparently anomalous results.
However, about 40 out of the 756 process unit costs shown at Appendix C appear to be significantly out of line with the other costs for the same processes. At this stage it has been possible to identify the following broad reasons why these costs are out of line:
the claimed level of activity for that process appears to be too high or low;
the cost pool allocated to that process appears to be too large or small;
the allocation of staff time to particular processes may be incorrect; and
other specific errors have been made in implementing the cost model.
Work is currently in progress to ensure the detailed validation of these figures and ensure the accuracy of the unit costs used for these extreme outliers in the settlement process itself. However, only 10 of the unit costs requiring detailed review relate to the most important 5 processes. This means that, while there will be some changes to allocations as these unit costs are refined, the orders of magnitude for MCCs will not change significantly.
In order to confirm this, the exemplifications have been rerun with these significantly outlying unit costs replaced by the average unit cost for each process of all those MCCs that lie within ± 1 Standard. The results of these exemplifications confirm that the impact of these adjustments to the major outliers on most MCCs is insignificant, amounting to less that 0.1% of their entitlements under both exemplification options. The only substantial exceptions to this are three MCCs with major outlier costs which receive between 1.5% and 4.1% in extra entitlement under Option 1 and between 1.4% and 1.9% under Option 2. This does reinforce the need to ensure that any issues regarding outlier costs have been properly resolved by the time of the settlement but it also demonstrates that the resolution of these issues will not fundamentally change the entitlements produced by the proposed mechanism.
The purpose of the paper is to demonstrate the behaviour of the mechanism rather than produce settlement numbers. The refinement of outlier unit costs will not fundamentally alter the way in which the mechanism performs even if it gives rise to marginal changes in allocations. This means that MCCs will still have a clear idea of how the implementation of the new mechanism will affect their allocations.
As a further part of these data validation activities a number of sensitivity analyses of the key variables that have been identified as important in the working paper responses have also been conducted. In particular, the impact of errors in the key data has been modelled to test their effect on the final allocations. The results of these sensitivity analyses are presented in Appendix H.
Although in theory there are a large number of possible exemplifications given the number of potential variables, the actual exemplifications have been limited to reflect the most realistic working assumptions. These are:
floors and ceilings are set to determine allowable process unit costs;
an option that allows each MCC to gain or lose a maximum of 3% from their last year's allocation and one that allows MCCs to lose a maximum of 3% but gain a maximum of 5%.
The allocations are reached on the basis of a series of steps which are shown in diagrammatic form at Appendix E. The results of the individual exemplifications are presented in Appendices F and G.
Although the main purpose of the paper is to illustrate the mechanism, it will help MCCs to know how all the elements of the allocation will be dealt with. Some of these elements fall outside the formula.
MCCs receive grant under the following broad heads:
A. general grant and top up bids (mostly not ringfenced)
B. money from the CJS Reserve for attrition, case progression officers and youth cases (ringfenced)
C. netted off money for enforcement (fines, community penalties and speeding) (ringfenced )
D. Libra, professional qualifications and PPP fees.
It is proposed that Groups A and B would be covered by the formula, but the amounts distributed in Group B would be ringfenced as before. This assumes funding under Group B continues to be provided from the Reserve. MCCs are familiar with the reasons why such money has to be ringfenced.
The sums in Group C are based on bids by MCCs. Because of the special circumstances of the netting off schemes, bidding as the basis for allocations will continue for next year. In making their business cases for this money, however, MCCs should have regard to the data they have collected on enforcement activities, as will the Court Service in analysing the business cases.
The sums in Group D are effectively fixed and will be paid outside the formula i.e. the amounts in question will be top-sliced from the total grant available.
MCCs are familiar with the existing 'hardship' arrangements by which they have been able to seek (1) 'top up' funding before the start of the financial year and (2) in some circumstances additional in-year funding. The rationale for the hardship provision was to help MCCs particularly affected by the continuation of the frozen formula. In practice, while MCCs have had to demonstrate need, funding has not always been rigorously tested against a hardship criterion. In principle, such funding was provided on a one-off basis.
MCCs' views are sought on whether these arrangements should continue. This paper proposes that the first type of provision should stop. The basic reason for hardship provision - absence of a working formula - will disappear if the formula is implemented. It may however be argued that the first application of the formula will be less exact, reflecting the underlying data, or that damping will prevent some MCCs from receiving their full requirement. The first reason, so far as it applies, will of course potentially affect all MCCs. Any limitation on damping will reflect a broad view on how much additional funding an MCC can reasonably absorb. In principle, therefore, it is not clear why MCCs should automatically be able to seek more money.
MCCs will bear in mind that any reserve set aside for top-up bidding or any other purpose is top-sliced from the overall grant and therefore reduces the amount available for general distribution.
It is separately proposed that, if funding pressures allow it, there should be an unallocated contingency reserve for genuine in-year difficulties that MCCs unexpectedly encounter. The test would be that the contingency was genuinely unforeseen and not reasonably foreseeable.
Next year's revenue grant will be affected by the outcome of the spending review. If extra money is available, Ministers will want to consider how to apply it alongside any movements of grant that the formula itself produces. At this stage no definite indication can be given. However priorities might include: improving the position of MCCs shown to be significantly underfunded; reducing the impact of the formula on those who might otherwise lose grant; and securing a contingency reserve. Ministers will need to allocate funding so as to give maximum impact to the Government's agenda for improving the performance of the criminal justice system.
In principle, once the spending review outcome is determined, the grant position for the next two years will be known. Some MCCs have previously argued for a 2 or even a 3 year settlement in order to provide reasonable certainty and stability and hence more favourable conditions for business planning. Such an approach also helps to cement the responsibility of MCCs for managing their budgets. On the other hand, it also increases the pressure on them to manage and prioritise expenditure when real events differ from planning assumptions.
While MCCs ought to know what the anticipated overall grant position is for the following 2 years, it is not clear that funding for individual MCCs can or should be fixed at this stage for 2004/05. This is because the impact of the new formula itself will need to be assessed. However MCCs' views are sought on whether a fixed 2 year settlement is feasible or desirable having regard to this consideration.
Implementation of the proposed formula needs to be recognised as only a first stage in a transitional process. Two factors are relevant here: first, the developmental process implicit in the introduction of the formula and, secondly, the potential impact of a unified courts agency as proposed by the White Paper 'Justice for All'.
Over time we should move away from depending on historical cost in particular. One of the stated aims for the new formula is that it should not produce perverse incentives, and it is therefore important, for example, that overhead costs do not remain sheltered from scrutiny as the analysis of local cost differentials progresses. Legitimate local variation such as regional wage pressures will, of course, need to be taken into account.
There are a number of factors that will need to be addressed in future developments to enable the movement from does cost to should cost, including:
development of best practice;
extension of this approach to cover the full range of costs;
improved forecasting of future activity levels;
review of quality of service measures and standards;
better understanding of local cost structures; and
allowing time to adjust to changing circumstances.
In future it is envisaged that the Court Service will facilitate arrangements to bring together service representatives to share experience and ideas on ways of working in MCCs. The process mapping information will provide a basis for these discussions. The work on the project to date has demonstrated a keen interest among MCCs in finding out more about how other MCCs work as well as sharing their own experiences of developing new ways of working. The impetus and opportunity provided by the current project will be built upon to ensure that this becomes a major contributor to continuous process improvement and development of a body of best practice.
The White Paper recommendations for a unified agency will take several years to implement, and MCCs will continue to be responsible for the efficient and effective management of magistrates' courts in the interim period. Pursuit of the developmental objectives described in the previous section remains legitimate and necessary during this period. Even when the new arrangements are in place, a mechanism for distributing funding to local units will be needed. The current formula and its development in the interim will provide a solid basis for future developments. As the unified administration takes shape clear understanding of the costs of delivering required outputs and the volume of work to be carried out will be necessary. This will enable an effective resource allocation and performance framework to be developed. So while the formula itself might be overtaken, the analysis and data which underpins it, together with improvements in the way in which resources are allocated to magistrates' courts work, will survive.
| Q.1 | Do you understand how the mechanism produced your notional allocation for 2002-03? |
| Q.2 | What is your view of the result? Give reasons. |
| Q.3 | Do you agree with the proposed treatment of the allocation, in terms of what is covered by and excluded from the formula (see paragraphs 44-48)? |
| Q.4 | Should the year-on-year limit for grant increases under the formula be 3% or 5%? Give reasons. |
| Q.5 | If the formula is implemented as proposed, should there be an opportunity for MCCs to 'top up' their allocation e.g. by a bidding process, before the start of financial year 2002-03? If so, what should the criteria be? What should the process be? |
| Q.6 | Should there be a separate reserve for in-year contingencies? What should the criteria be? What should the process be? |
| Q.7 | Is it feasible to provide a 2 year (or longer) settlement? |